Abstract Sleep abnormalities commonly occur among chronic cocaine users after withdrawal, including loss of total sleep time and increase in sleep fragmentation. The withdrawal-associated sleep problems are speculated to foster cocaine use and relapse, however, whether and how sleep mechanisms may regulate the brain reward circuitry and impact relapse-like behaviors remain elusive. Using a rat cocaine self-administration model to recapitulate sleep loss and fragmentation after withdrawal, the Huang lab has obtained direct evidence of sleep-induced regulation of cocaine seeking: experimentally increasing REM (without changing NREM) sleep episode durations reduces cocaine seeking after withdrawal, suggesting REM sleep-associated mechanisms in this regulation. Neural mechanism studies have focused on the nucleus accumbens (NAc), a key brain region for reward processing. Progressive accumulation of the GluA1-rich, calcium-permeable, AMPA receptors (CP- AMPARs) at synapses in the NAc critically contributes to the intensified cocaine seeking after withdrawal. Importantly, behavioral sleep interventions that increase REM sleep episode durations lead to decreased accumulation of NAc synaptic CP-AMPARs. These results not only suggest NAc CP-AMPARs as key neuronal substrates that express REM sleep-induced anti-relapse effects, but raise the critical question ? how do REM sleep interventions route to NAc CP-AMPARs? Published and preliminary results suggest that the melanin- concentrating hormone (MCH) neurons in the lateral hypothalamus and zona incerta (LH for short) may play an important role in the REM sleep-induced anti-relapse effects. LH MCH neurons predominantly fire during REM sleep. Behaviorally induced sleep rebound after REM sleep restriction, a strategy utilized by our sleep intervention, further enhances the activity of these neurons. Moreover, MCH neurons project to the NAc, where MCH receptors (MCHRs) are highly expressed; MCHR signaling in the NAc strongly regulates the phosphorylation and facilitates synaptic removal of GluA1-containing AMPARs. Preliminary results further show that LH MCH neurons exhibited reduced membrane excitability after withdrawal from cocaine, whereas mimicking MCH release by intra-NAc infusion of MCH during light (sleep) phase led to reduction of synaptic CP-AMPARs in the NAc and decreased withdrawal-associated cocaine seeking. Together, these results suggest that REM sleep interventions may engage LH MCH neural activity to produce anti-relapse effects after withdrawal. This application will test the hypothesis that MCH signaling during sleep contributes to REM sleep-induced anti-relapse effects after withdrawal from cocaine. In contrast to the current practice focusing on NREM sleep, this proposal will identify a novel REM sleep mechanism that produces anti-relapse effects. Moreover, the proposal emphasizes target manipulations during sleep rather than in wakefulness. Concerning the high comorbidity between drug withdrawal and sleep disturbance, this application reveals novel strategies for developing sleep-based therapies for drug addiction, thus is highly relevant to NIH?s mission.